Hill sprints may not build the strength you think

Uphill running is a staple of athletic training, used on the assumption that fighting a gradient builds lower-limb strength, jumping power, running ability and aerobic fitness all at once. Yet the randomised trials testing those assumptions have reported inconsistent findings, and no previous review had focused exclusively on uphill running across these performance domains. That leaves coaches and recreational runners prescribing hill work on tradition and plausibility rather than on pooled evidence. This systematic review and meta-analysis set out to establish what structured uphill running actually delivers in athletes and recreationally active adults.

The reviewers searched six databases — CINAHL, Embase, PubMed, Scopus, SPORTDiscus and Web of Science — up to June 2025 for randomised controlled trials comparing uphill running against downhill running, level-ground running or usual training. Thirteen RCTs published between 2006 and 2023 met the criteria, covering 308 participants, with a mean Physiotherapy Evidence Database quality score of 5.4. Seven trials provided enough data for quantitative pooling. Typical protocols involved 10 to 14 uphill sprints of 30 seconds at roughly a 10% gradient, performed 2 to 3 times weekly over 2 to 8 weeks.

The pooled analyses came up empty on all three fronts: no significant effect on isokinetic knee strength (SMD 0.09), jumping performance (SMD 0.15) or aerobic capacity (SMD 0.31), each with confidence intervals crossing zero and no statistical heterogeneity. The narrative synthesis was more encouraging, describing improvements in running economy, repeated sprint ability, ventilatory threshold and lactate clearance in individual trials. However, the small number of trials contributing to each pooled comparison, small sample sizes and variable intervention protocols left the overall certainty of the evidence low.

This is the first review to isolate uphill running rather than fold it into broader sprint or interval training literature, and its message complicates a widespread training belief. It does not show that hill running is useless — the narrative findings on running economy and repeated sprints point in a favourable direction — but it does show that the strength, jump and aerobic-capacity benefits often claimed for it are not confirmed when trials are pooled. Open questions remain about optimal gradients, longer training durations and whether adequately powered trials would detect effects the current small studies could not.

If you run hills expecting them to double as a leg-strength or jump-training session, this is a reason to temper that expectation: the pooled trial evidence does not currently support those effects. The more defensible case for hill sprints, on this review's reading, lies in running-specific qualities such as economy and repeated sprint ability, though even those rest on narrative rather than pooled findings. It is also a reminder that a popular training method can persist for decades on plausibility alone. Better-standardised, larger trials are needed before firm conclusions in either direction.